ABSTRACT
Topological thinning can be used to accurately identify the central path through a computer model of the colon generated using computed tomography colonography. The central path can subsequently be used to simplify the task of navigation within the colon model. Unfortunately standard topological thinning is an extremely inefficient process. We present an optimised version of topological thinning that significantly improves the performance of centreline calculation without compromising the accuracy of the result. This is achieved by using lookup tables to reduce the computational burden associated with the thinning process.
Subject(s)
Colonography, Computed Tomographic , Imaging, Three-Dimensional , Radiographic Image Interpretation, Computer-Assisted , Endoscopy , Humans , IrelandABSTRACT
A free visual programming-based image analysis development environment for medical imaging applications called NeatVision was developed to provide high-level access to a wide range of image processing algorithms through a well-defined, easy-to-use graphical interface. The system contains over 300 image manipulation, processing, and analysis algorithms. For more advanced users, an upgrade path is provided to extend the core library with use of the developer's interface, giving users access to additional plug-in features, automatic source code generation, compilation with full error feedback, and dynamic algorithm updates. NeatVision was designed to allow users at all levels of expertise to focus on the computer vision design task for computer-aided diagnostic (CAD) applications rather than the subtleties of a particular programming language. The environment allows the designers of image analysis-based CAD techniques to implement their ideas in a dynamic and straightforward manner. Both NeatVision standard and developer's versions can be downloaded free of charge from the Internet and can run on a variety of computer platforms.
Subject(s)
Diagnosis, Computer-Assisted/methods , Imaging, Three-Dimensional , Radiography/methods , Radiology/methods , Software , Medical InformaticsABSTRACT
Computer-aided analysis of medical images usually involves the development of custom software applications that interpret, process, and ultimately display medical image data. The interpretation stage involves decoding the image data and presenting them to the application developer for further processing. A toolkit has been created specifically for interpreting medical image data; it thus acts as a platform for development of medical imaging applications. The toolkit, which is referred to as NeatMed, is intended to reduce development time by eliminating the need for the application developer to deal directly with medical image data. NeatMed was implemented by using Java, a programming language with a range of attractive features including ease of use, extensive support material, and portability. NeatMed was developed specifically for use in a research environment. Straightforward to use and well documented, it is intended as an alternative to commercially available medical imaging toolkits. NeatMed currently provides support for the Digital Imaging and Communications in Medicine and Analyze medical image file formats. Support material including sample source code is available via the Internet; links to related resources are also provided. Most important, NeatMed is freely available and its continuing development is motivated by requests and suggestions from end users.
